Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    collection of plural nouns or actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ologi
    0.47
    swith
    0.39
    ologija
    0.39
    ología
    0.37
    TRR
    0.37
    schen
    0.37
    ningen
    0.37
    mez
    0.36
    TRAN
    0.36
    anyl
    0.36
    POSITIVE LOGITS
    heets
    1.21
    heet
    1.11
    ystem
    1.09
    ervice
    1.07
    chool
    1.04
    ource
    1.03
    ervices
    1.02
    hift
    1.02
    hoot
    1.01
    ociety
    1.00
    Act Density 0.048%

    No Known Activations